NORTH- HOLLAND
The Past and Future of Constructive
Technology Assessment
J O H A N S C H O T and A R I E RIP
ABSTRACT
Constructive technology assessment (C-f A) is a member of the family of technology assessment approaches,
developed in particular in the Netherlands and Denmark. CTA shifts the focus away from assessing impacts
of new technologies to broadening design, development, and implementation processes. Explicit CTA has
concentrated on dialogue among and early interaction with new actors. The idea has been taken up by
actors other than governments (consumers, producers). CTA implies a modulation of ongoing technological
developments, and an understanding of the dynamics of such modulation is used to identify and briefly discuss
three generic strategies for CTA: technology forcing, strategic niche management, and loci for alignmenl.
Modulation activities are to be located in the broader issue of how our societies handle new technology at all.
The established division of labor between promotion and control should be mitigated by sociotechnical criticism.
This underlines the need for reflection on role and value profile of CTA agents. © 1997 Elsevier Science Inc.
Introduction
In the three decades since the first articulation of technology assessment (TA), a
whole family of TA approaches has emerged. Smits, Leyten, and Den Hertog [1]
distinguish awareness TA, strategic TA, and constructive TA. Grin and van der Graaf
[2] emphasize interactive TA, a more symmetrical version of what used to be called
participatory T A [3]. This family of approaches is characterized by its commitment to
what we see as an overall T A philosophy: to reduce the human costs of trial and error
learning in society's handling of new technologies, and to do so by anticipating potential
impacts and feeding these insights back into decision making, and into actors' strategies.
One member of this family of approaches is constructive TA (CTA), which originated in the mid-1980s in the Netherlands. In the 1984 Policy Memorandum on T A [4] the
interest in constructive approaches is clear, and the term constructive is used regularly. In
1987, the Netherlands Organisation of Technology Assessment (NOTA) published a
background study on CTA [5]. Since then, policy makers, commentators, and researchers
in various countries have taken up the concept of CTA and further articulated it. As
ARIE RIP is Professor at the Department of Philosophy of Science and Technology at the University of
Twente, Enschede, The Netherlands.
J O H A N SCHOT is Assistant Professor at the Department of Philosophy of Science and Technology at
the University of Twente, Enschede, The Netherlands.
Address correspondence to Johan Schot and Arie Rip, Centre for Studies of Science, Technology and
Society; University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands.
Technological Forecasting and Social Change 54, 251-268 (1996)
© 1997 Elsevier Science Inc.
655 A v e n u e of the Americas, New York, NY 10010
0040-1625/97/$17.00
PII S0040-1625(96)00180-I
252
J. S C H O T A N D A. RIP
a result, CTA has a diffuse and emerging character, allowing each participant in the
articulation process to emphasize different aspects. At the same time, however, there
is a general recognition of the overall thrust of CTA: to broaden the design of new
technologies (and the redesign of old technologies). Feedback of TA activities into the
actual construction of technology is crucial, and strategies and tools contributing to
such feedback make up CTA5 Such strategies and tools can range from dialogue
workshops and social experiments, to technology forcing programs and platforms. One
can now speak of a paradigm of CTA to indicate the combination of the widely shared
diagnosis of the need for broadening of technical design and the exemplary experiences
available [6].
The aim of this article is to present and evaluate what has happened already under
the label of CTA and to discuss the future of CTA. In the second section, we describe
where, how, and why the idea of CTA was formulated and how it was taken up. In the
third section we evaluate how far CTA has evolved. The future of CTA is addressed
in two ways: in the fourth section we discuss three generic strategices to implement
CTA; in the fifth section we argue that CTA (and TA generally) is never just a tool
or a particular kind of policy analysis. It is always also part of the "politics" of managing
technology in society. Managing must be taken here in a broad sense, referring to
patterns in the actions of actors responding to an evolving situation. This makes it clear
that CTA analysis and action must reflect on its wider implications, and in the fifth
section we use the Luddite actions against textile machines in the early 19th century
as an example. Luddite actions are often misperceived as a conservative reaction to
technical innovations. We point out their intent to integrate societal concerns in technical
change and argue that their form of socio-technical criticism is reincarnated in constructive TA. In the sixth and last section we address the question of who decides which
ways of handling technologies are best, and suggest that CTA agents might do well to
develop a competence in irony.
The Past of Constructive T A
The Dutch policy Memorandum of 1984 [4] positioned itself as being part of a new
phase of TA, where T A is linked to decision making (as exemplified by the approach
of the U.S. Office of Technology Assessment [OTA]) and is also embedded in broader
political and societal processes of articulation of opinions and decisions about scientific
and technological innovation. Accordingly, the Memorandum formulated an overall
goal for this new T A (the label CTA was not used in the Memorandum, nor any other):
broader decision making about science and technology in society. "Broader" must be
taken here as referring to the substantial criteria taken into account in technological
development, as well as referring to the groups, organizations and institutions involved.
The Memorandum argued that the function of TA studies should be to let societal
aspects become additional design criteria.
In the Netherlands, one result of the Memorandum was the establishment, in
1986, of the Netherlands Organisation of Technology Assessment (NOTA), now called
The emphasis on construction explains why the development of C'TA has been guided by (and was
closely linked to) the development of technologies studies. This is far less the case for other branches of the
TA family, that have been influenced by other disciplines and interdisciplinary fields. CTA needed a theory
on technology dynamics for identifying feedback mechanisms. In addition, research on CTA (feedback processes
and use of tools) will help sharpen insights about technology dynamics.
CONSTRUCTIVE TECHNOLOGY ASSESSMENT
253
Rathenau Institute. 2 N O T A developed its ideas on new and broader T A through several
projects, evaluations, and studies. In retrospect, it is possible to see two paths emerging.
One path, becoming dominant, focused on stimulating discussions and analysis that
would contribute to a social debate and the articulation of political opinion (in Parliament) [7]. The Rathenau Institute used several methods for organizing a broader debate.
among others the public debate based on the Danish model of consensus conferences
[8]. In a public debate, a panel of citizens, recruited through advertisements in the
newspapers, discuss a specific subject with experts in a well-prepared two- or three-day
meeting. The outcome is a statement made by the citizen panel containing its judgments
and questions for further development. The experts have the opportunity to comment
on the final statement. According to the Rathenau Institute. these debates serve a dual
purpose: (1) "to introduce the viewpoints of citizens in the process of political opinion
forming"; and (2) "to stimulate interest in and discussion on a subject concerning society
as a whole" ([9], page 26).
A second path N O T A took is the development of CTA. Among N O T A ' s activities
oriented explicitly toward CTA, a first step was a background study on CTA [5]. This
study was important in identifying opportunities for CTA, in highlighting the importance
of what was called "societal learning how to handle new technology," and in offering
concrete recommendations. One recommendation, to devote at least 1% of the funding
in every technological innovation stimulation program to T A studies, was adopted in
a few cases, for example, in the stimulation program for biotechnology. The results of
this first study influenced several N O T A projects. It resulted, for example, in an emphasis
on including technology developers and on articulating design criteria in various projects.
In its report, "Technology Assessment: To Adjust or to Channel," published after three
years of work in order to present N O T A ' s approach, the section on telecommunication
T A activities finished as follows: "this programme illustrated that it is quite feasible, and
indeed pertinent, that developers of technology enter discussions with other concerned
parties in society during the actual designing process and by doing so contribute toward
the further development and introduction of the relevant technology" ([10], page 4).
Already in the 1984 Memorandum, the importance of demonstration projects in
which construction principles and design criteria are developed was recognized ([4],
page 22). This led to a major project set up to stimulate the introduction of environmentally friendly technologies in firms and other organizations, the so-called PRISMA
project [11].
In parallel, N O T A continued to invest in further work to articulate the CTA
perspective. In 1990, a study was done on how CTA could profit from insights from
technology studies [12]. In 1991, an international workshop was supported (together
with the Dutch Ministry of Economic Affairs) resulting in a 1995 book publication [6].
In 1995, a study of CTA methodology for biotechnology firms was commissioned [13].
The results of these studies guide our following evaluation of CTA.
CTA was taken up by various other organizations in the Netherlands. For example,
the institute for consumer research (SWOKA) developed a procedure, called Future
Images for Consumers, for incorporating consumers and their wishes into design processes [14]. The procedure consists of a process of iterative meetings in which room is
created for discussions and negotiations on design criteria and visions of both producers
The other main effect of the Memorandum was the establishment of a Foundation for Public Understanding of Science and Technology, to develop information and public understanding activities itself, as well as
fund such activities by others.
254
J. SCHOT AND A. RIP
and consumers. The method has been tried in such cases as a project commissioned by
the Dutch Program on Sustainable Technologies (DTO) on the introduction of novel
protein foods which could replace meat in the diet. In a final declaration issued after
a third meeting, design requirements for such new foods were stipulated, and market
opportunities and possible actions were defined [15]. The D T O program will continue
to incorporate CTA elements in its activities as part of its attempts to set up so-called
illustration processes of new sustainable technologies [16].
The idea of CTA also influenced the development of technology policy in the
Netherlands ([17], page 19; [10], page 11). In several documents, the Ministry of Economic Affairs highlighted the need for a third phase in technology policy that would
focus upon improving the societal embedding of new technologies, thus complementing
the supply policies emphasized in the first phase and the diffusion policies of the second
phase. The newly proposed policy was implemented through: (1) the organization of
strategic conferences with a broad set of constituencies discussing new priorities and
leading to demand articulation and design criteria; and (2) the stimulation of experiments
with new technologies in which citizens participate. An example is Digital City Amsterdam, a project in which citizens experiment with e-mail and Internet, build user competence, and provide input for further technology development. A series of education
initiatives also form part of the policy [18].
CTA activities were not developed only in the Netherlands. In Denmark, such
activities emerged in the late 1980s when the overall shift to more proactive approaches
took up the challenge of feeding TA insights back into technological development and
adoption. The form these activities took relates to the Danish political culture where
public involvement has always been considered important. Thus, there was and is an
emphasis on awareness initiatives [8], on social experiments (e.g., with information and
communication technology) [3], and on consensus conferences and dialogue workshops
(organized among others by the Danish TA organization) [19, 20]. Most of these activities
are not labeled as CTA by their organizers. However, some Danish evaluators of these
activities have found this label useful to position the changes that occurred in Danish T A
practice because the overall thrust of these activities is aimed at influencing technological
design and implementation.
In other countries, CTA activities are present but not labeled as such. Examples
can be found in Norway [21] and Germany [22]. In addition, CTA is now discussed
in technology policy documents. In an Organization for Economic Cooperation and
Development (OECD) report, the emphasis on investment in new technologies is complemented by an argument for a more subtle approach to technology assessment [23].
Under the heading, "towards a broad-based consensus: the role of constructive technology assessment," O E C D considers the role of the state in counteracting externalities of
technological change and recommends experimentation with new institutional structures
and arrangements (Section J of the Conclusions and Recommendations). For OECD,
the term "constructive" indicates the expectation of minimizing mismatches, wrong
investments, and possible social conflict, which one can read as a version of our formulation of constructive technology assessment.
In the new R&D programs of the European Union (the 4th Framework Program
and the 5th Program in the making), strengthening links with users and an orientation
of R & D toward social conditions for innovation can be seen [1]. The same trend is
visible in the 1992 report of the Carnegie Commission in the United States ([24];
see also [25], pages 21-26, 167-201). The limitations of supply-driven initiatives are
recognized, while the report calls for more interaction between supply and demand.
CONSTRUCTIVE TECHNOLOGY ASSESSMENT
255
Attention is paid to the capacity of society to incorporate technological changes and
the need to articulate demand and acceptability. The point is made even more forcefully
by the economist Soete [26] when he argues that technology policies aiming at promotion
and CTA are two sides of the same coin--a strategy for achieving goals of wealth
creation, sustainable development, safety, and quality of life. Therefore, as we have
argued elsewhere, CTA can thus be seen as part of a redefinition of the boundary between
current technology policies and technology assessment [6]. In an action perspective,
both aim at promoting technologies that promise to have many desirable and few
undesirable impacts.
CTA Revisited
How far has CTA evolved? It is not a clearly defined area, as we noted above,
and it need not, perhaps, become one. CTA is not simply a management tool. CTA
can be seen as a new design practice (which includes tools) in which impacts are
anticipated, users and other impacted communities are involved from the start and in
an interactive way, and which contains an element of societal learning. This perspective
allows us to identify three achievements, and their limitations, of CTA, and to develop
more systematically the overall thrust of CTA and what this implies for the future.
Thus, we combine a retrospective evaluation of CTA with an articulation of its further
potential. It is the latter which we discuss in the next section, when we present generic
strategies for CTA.
THREE ACHIEVEMENTS OF CTA, AND THEIR LIMITATIONS
Several methods have been developed that, while not exclusive, were particularly
suited to CTA. In every T A study, technological developments as well as societal issues,
views and preferences need to be mapped. Such "social mapping" [27] is an extension
of stakeholder analysis as practiced by OTA. For CTA, it can and should be further
extended by mapping the dynamics of technological developments--sociotechnical mapping for short [28]. Other methods relate to anticipatory agenda building, with the
Future Images of Consumers Procedure about new technologies, developed by SWOKA
and dialogue workshops introduced in Denmark, as two interesting examples.
Methods for early and controlled experimentation with new technologies are increasingly recognized as important, and applied. Such experiments can be seen as
forceful sociotechnical demonstrators. They allow the translation of broader societal
scenarios and agendas into actual design criteria and other orienters of technological
development. Sociotechnical demonstrators embody visions concretely, and the experience with them allows modification of the vision and the technology in one movement.
The potential of these methods can be further exploited if the aspect of (societal)
learning, characteristic for CTA, is taken up explicitly.
Full-fledged CTA activities in the Netherlands and Denmark, as well as some of
the recent technology policy declarations, emphasize dialogue and the articulation of
demand and acceptability. This is an important and necessary part of CTA. It leads to
(diffuse) societal agenda building for new technology. However, it turns out to be
difficult to organize feedback of such societal agendas into actual technological development. There often is no special effort at feedback other than publicizing results and
hoping that technology actors will respond. There may well be effects of such exercises,
however, for example through diffuse credibility pressure (see the next section), or
256
J. SCHOT AND A. RIP
when technology actors have a (legitimation) problem that they cannot solve in a usual
way (as happened in modern biotechnology).
Semi-governmental (C)TA agents like the Dutch Rathenau Institute and the Danish
Board of Technology are well placed to create spaces for interaction and dialogue, and
their activities have shown the usefulness of such exercises. Clearly, to come closer to
the original goal of achieving better technologies (in a better society), a concerted
attempt at feedback into decision making, and strategies of technology actors and other
forms of leverage are an important next step. We see this as the future of CTA,
and we will develop generic strategies to influence technological developments in the
next section.
Interestingly, within the world of technological development, and also with consumer and social pressure groups, de facto CTA activities, in the sense of attempts to
broaden design and implementation processes, are increasing. The attempts at concurrent (or simultaneous) engineering, where product development, production, implementation, and marketing are worked on in parallel, instead of sequentially, are sometimes
extended to include work on waste management, social acceptance, and adoption. This
happens emphatically in biotechnological firms which have encountered legitimation
and acceptance problems that they could not solve by traditional means [13, 29]. Environmental pressures have led to so-called integral chain management with its ambition to
close material cycles. Accordingly, waste handling activities need to be connected to
design and production activities [30]. New networks have been established, such as the
European Partners for the Environment, in which NGOs, businesses, governments, and
researchers work together in sustainability laboratories to develop new products and
processes and define design criteria. Consumers and pressure groups have become
interested in building "nexusses" between technological worlds and society [14, 31].
Increasingly, they are asked to participate in "platforms" (established for the introduction of new technology such as high definition TV) and to discuss technological and
product options with firms.
What is becoming clear is that there are more CTA agents than the governmental,
parliamentarian, and semi-governmental bodies that were the main CTA agents until
recently, and they are still the most visible ones. Firms, consumer organizations, and
other actors can anticipate and work toward feedback into technological developments,
and may well be able to do so more effectively than governmental agents. Thus, it is
important to recognize these developments and introduce consumer CTA (cCTA) [14],
producer CTA (pCTA) [13], etc. and elaborate the specifics of these approaches.
An additional point is that responsibility for "managing" technology in society, as
we call it [6], is not limited to governmental actors. W e can develop this point a little
further, using a distinction introduced in the background study [5]. Feedback into
technological developments is an activity where three kinds of actors are involved.
"Technology actors" are those who invest in, and carry, technological developments,
and are always the target of CTA activities, Examples of technology actors are firms,
some government agencies, national laboratories, and technology programs. Second are
"societal actors" who anticipate and try to feed back into technological developments
(through regulation, campaigning, educating, etc.). Examples are government agencies
and various societal groups, but firms and other technology actors can play this role
as well, and thus short-circuit the feedback. Third, there are actors at a meta level.
Governments adjudicating among actors are an obvious example. Without the authority
of government agencies, meta-level activity still occurs, but it is better described as
facilitation and modulation of the interaction between technology actors and societal
CONSTRUCTIVE TECHNOLOGY ASSESSMENT
257
actors. T A institutions could take on this role, but so could government agencies or
institutions without an explicit T A mandate.
This implies not only that responsibility for "managing" technology in society is
distributed over more actors, but also that the nature of the responsibility is different
for different kinds of actors. In a sense, all three kinds of actors are necessary.
THE THRUST OF CTA AND ITS FURTHER POTENTIAL
CTA is built around the attempt to anticipate effects or impacts of new technologies
or new projects with a strong technological component. This is a core component of
any T A effort. In traditional TA, the technology or the project is taken as given, and
thus seen as a static entity. For CTA, the dynamics of the process are central, and
impacts are viewed as being built up, and co-produced, during the process of technical
change. Many technology studies have shown that impacts are not just passive effects
of a given technology on its environment, but are actively sought (or avoided) by
technology producers, users, and third actors such as governments, unions, and pressure
groups alike [32, 33, 53]. These actors are co-producers of impacts. (This should not be
read to imply that we believe all actors have equal access and power in various stages
of technology development.)
Co-production processes include anticipation. Technical change is driven partly by
the historical experience of actors, their views of the future, and their perceptions of
the promise or threat of impacts which will change over time. In turn, technical change
generates new impacts when applied to new social settings. These dynamic, multi-actor,
and decentralized co-production processes are shot through with assessments. Thus,
the situation is not one where T A has to introduce assessment. Assessment occurs all
the time, and it is a modulation of ongoing processes of assessment (and feedback)
which is in order. This, we claim, is the thrust of CTA.
Modulation of ongoing processes is an empty phrase if one does not specify the
goals or criteria that guide modulation activities. Actors have concrete goals, interests,
and values. But CTA, like T A in general, should not be partisan and identify with
particular actors' goals and interests. Thus, meta-level criteria for desirable development
should be specified [34]. In addition to the core idea of all T A approaches--that it is
better to anticipate what happens--there are two other meta-level criteria.
Learning must occur, and it should be broad and deep. Broad learning entails an
exploration of possible new linkages between a range of aspects such as design options,
user demands, and issues of political and societal acceptability. With deep learning we
indicate that both first-order learning and second-order learning occur. In first-order
learning, one improves one's working toward given goals, while second-order learning
requires clarifying values and ways of relating values to each other, succinctly called
collective value learning by Wynne [35].
Reflexivity is also needed, and for two reasons. First, reflexivity about technology
is necessary to avoid falling back in a naive contrast between technology and society,
and its attendant proponents-opponents dichotomies [36]. Actors need to be reflexive
about co-production of technology and its effects, because this creates opportunities to
improve the process. Second, reflexivity is necessary to recognize the different roles of
actors in technological development and the need to have all three kinds of actors
present.
Reflexivity and learning do not come easily, but even anticipation is more difficult
than the ideology of prudence would suggest. Many studies have shown that technology
actors do not use the results of T A studies, or even the results of marketing studies
258
J. SCHOT AND A. RIP
[29, 37]. If impacts, even if they are anticipated, are not taken into account in actual
decision making, special efforts are necessary. It is the occurrence of such efforts that
qualifies the development as a desirable development.
Independent of these issues of decision making, we argue that properties of a
technology, artifact, or system ar not given beforehand, but they co-evolve with the
interactions occurring during development, implementation, adoption, and wider use.
Rip and Kemp [33] summarize our understanding of the mechanisms as follows. The
co-evolution processes start with novel options against a background of existing technological and societal regimes. With increasing investments (material, social, symbolic),
irreversibilities such as sunk investments, lock-ins, and path dependencies emerge. There
are patterns in the irreversibilities, but no a priori laws. Interactive processes determine
outcomes, and no one actor can be singled out as exclusively responsible. Inversely, no
one actor can force the process in the direction he or she desires. Command and control
approaches do not work, and modulation is the best one can do.
Some path dependencies are better than others, even if one cannot specify beforehand which ones will turn out to be better. CTA is an attempt to improve our chances
to arrive at better path dependencies by broadening technological design and development. The achievements so far, however, are not enough. Societal agenda building has
been achieved, but feedback into technological development is not assured. Therefore,
it is necessary to rethink the approach of CTA on the basis of our recent understanding
of technology dynamics.
Basically, what we want to do is to develop "generic strategies" for CTA. We call
the strategies generic because they can be taken up by all actors (although not in the
same way). Each actor has its own limitations and opportunities to implement these
CTA strategies. In such a multi-actor and decentralized situation, the notion of strategy
takes a different complexion. Strategy does not refer to the intentional, almost military
approach that boards of firms and some governments appear to embrace. It is like
Mintzberg's [38] "pattern strategy": the pattern in the actions of actors responding to
an evolving situation, which may become reflexive and consciously worked for.
The Future of CTA: CTA Strategies
Recent insights about technology dynamics allow us to identify three generic CTA
strategies: technology forcing, strategic niche management, and stimulation (or creation)
of alignment. These strategies have been discussed elsewhere in a preliminary way [33,
39, 40]. We discuss them separately, but note that a mix of the three should be sought
for in each particular case in order to stimulate reflexivity, anticipation, and learning.
INVERSE ANTICIPATION AND FEEDBACK: T E C H N O L O G Y FORCING
Technology forcing, as an intentional effect of government regulation, seems an
attractive option to link science and technology to societal goals. It is the inverse of
traditional TA, which takes the technology as given and explores potential impacts. In
technology forcing by regulation, desired impacts are stipulated; for example, levels of
pollutants in motor car exhaust gases, and technology actors are challenged to come
up with technologies that fulfill these requirements. While the term technology forcing
was originally used to indicate an effect of government regulation, the approach can
be used more broadly.
A good example of technology forcing by a government is the clean air standards
set in California in 1988. The state prescribed that by 1998 2% of car sales must be
C O N S T R U C T I V E T E C H N O L O G Y ASSESSMENT
259
zero-emission vehicles. While this mandate does not prescribe a specific technological
solution, it has become clear that only the electric vehicle will be able to meet the
standard in 1998. Government regulation in fact induced a broadening of the existing
R & D and technological development activities to include new requirements, in this
case the design and production of a zero-emission vehicle. Note that it is not necessary
to work through regulation; a procurement program that creates an assured market for
a new technology could also have a technology forcing effect. For electric vehicles, this
route has been followed in Sweden. 3
In general terms, the strategy of technology forcing is to prescribe specifications
to be achieved in an authoritative manner. Then, the required technology will somehow
be developed. There is a general problem, however: how can a government anticipate
what are realistic requirements on technology yet to be developed? Actors may well
contest the feasibility of the requirements, and use this as an argument not to work in
the desired direction. In addition, in many cases technology forcing is not possible
because governments are not in the position and/or not willing or able to formulate
stringent standards due to lobbying efforts of pressure groups, lack of societal backing,
and other barriers. For example, California standards cannot be formulated by national
governments in the European Union because they would be considered barriers to
free trade.
While the approach of technology forcing is clear, the actual implementation is
beset with problems. The linear dream, of wishes and requirements leading to desired,
well-functioning technology, is just that--a dream. It is important to analyze the nature
of the difficulties in order to develop technology forcing into a realistic CTA approach.
At the "forcing" side there is an actor (most often a government agency) wanting
to get functions realized, at some distance from the actors who should do the realizing.
That is, it must be seen as a strategy of a principal or purported principal trying to get
agents to do the work. Such a strategy runs into trouble because of lack of information
with the principal, lack of power and/or understanding of possibilities with the principal,
and counter-strategies of the agents building on these limitations of the principal. The
Clean Air Act of the 1970s foundered because of these reasons. The automobile industry
argued that they could not meet the standards, and turned this into a self-fulfilling
prophecy--with congressional support--when the introduction of the standards was
postponed several times [41, 42]. The clean car policy of California in the 1990s has
faced the same difficulties, and its goals have also been postponed.
Such delays need not be viewed as only strategic undermining of the right thing
to do. They allow repair work to be done which is often needed in cases of moving
targets. Technology forcing is played out in and through strategic games, and it actually
becomes effective that way. Analysts have argued the governments should act accordingly and follow a "fail-soft strategy" [28, 42]. Governments must be strict, and must
be seen to be strict in order to create a technology forcing effect. At the same time,
however, if industry shows good faith efforts and still fails to meet the standards, they
must not be punished for noncompliance. This is not, however, a simple recipe to resolve
the dilemma of how to keep the regulation pressure high while at the same time ensuring
constructive reactions from industry. Technology forcers might want to follow the
The Swedish National Board for Industrial and Technical Development (NUTEK) was instrumental in
developing a purchasing program through which various fleet operators have formed a consortium and specified
product requirements. The consortium invited producers to bid and offered in return a guaranteed sale if they
were able to meet the requirements.
260
J. SCHOT AND A. RIP
suggestion of Schot, Hoogma, and Elzen [40] and combine technology forcing with
strategies of niche management and network building.
Even without the struggle between the self-styled principal and the technology
actors, the complexities of the technology part of the strategy of technology forcing are
sufficient to make a linear goal-setting and implementation approach unrealistic. The
technology to be realized depends on sociotechnical linkages (existing and to be created),
on path dependencies, and on the peculiarities of adoption and diffusion which determine
the actual functioning of a technology. In addition, impacts are dynamic: they are
elaborated and evolve across the course of lengthy development and implementation
projects. Therefore, even if clear values are present and shared, it is often impossible
to identify an optimum strategy beforehand. Thus, even with good faith efforts from
everyone involved, success will fail to materialize if the heterogeneity, the historical
character of technological development, and the need for experimentation and societal
learning are not taken into account [43].
For CTA, it is important to recognize that the strategy of inverse anticipation, that
is, technology forcing in the broad sense can also be applied by other actors such as
banks, insurance companies, standard-setting bodies, user-firms, and social pressure
groups. Of course, such actors have no (or occasionally only delegated) authority to
make technology forcing standards mandatory. Still, in particular cases, their action
might have substantial impacts, directly or because of actual or potential take up by
the government. Insurance companies could prescribe certain technologies in order to
limit their risks. Government could be instrumental by developing a strict liability regime
for risks of products and processes. (Note that such a regime might also lead to a riskaverse product and process development policy within firms.) Manufacturers using
chemicals could impose demands on their suppliers. For example, firms within the foodpackaging industry are asking for plastic without PVC; car manufacturers are demanding
plastics that can be recycled more easily. Government agencies can take measures to
encourage this, for example by making producers responsible for take-back of their
products, as is happening in several European countries. Credibility pressure can be
reinforced by governments, for example through forcing reporting requirements on
firms. The Bhopal explosion in 1984 prompted the U.S. Congress to enact the Emergency
Planning and Community Right to Know Act in 1986, requiring facilities to disclose
chemical information to the wider public. This caused many firms to implement new
environmental and safety technologies, to achieve greater safety, and to be able to tell
a better story to the public.
How is technology forcing evaluated from a CTA perspective? It is a way to
broaden design, directly (if technology forcing dilemmas are recognized and worked
upon) and indirectly through anticipatory and forceful agenda building. This relates to
the process of technology forcing, when it leads to a number of iterative explorations
of potential impacts of a range of technologies, systems or regimes, and assessment of
their desirability. These explorations will influence and modify expectations that guide
ongoing technological development. The California zero-emission standard, for example,
changed existing expectations about the future of the electric vehicle and eventually
led to higher investments in both electric vehicles and other alternatives for reducing
emissions substantially. The California initiative turned the electric vehicle from a toy
and R&D project with no future, into a marketable product. Around this product a
speculative market of early promises and expectations emerged. In this market, expectations about a range of other alternatives (including hybrids, improved gasoline cars,
natural gas vehicles) also became better articulated (for more details see [38]). Due to the
CONSTRUCTIVE TECHNOLOGY ASSESSMENT
261
rising importance of this market, electric vehicle design requirements were negotiated,
included better specifications, were more broadly shared, and were backed by studies
and test results.
Thus, through a process of assessments, technology actors are better able to anticipate and accordingly shape their R & D and introduction strategies. Use of tools such
as dialogue workshops can be of help. To maintain the characteristic orchestrating
aspect of this generic strategy, the forcing aspect should not disappear altogether.
GRADED LEARNING AND FEEDBACK: STRATEGIC NICHE MANAGEMENT
In technology forcing, there is a distance between the forcing actor and the technology developers, and this creates problems for modulating and broadening design processes. One can start at the other end, with the technology developers, and ask how
processes of development and introduction can be broadened from the start. We again
use government experience, now of agencies developing desired technologies themselves, or having them developed under their direct authority (where the developer is
thus directly linked to political authority and accountability), as a stepping stone to
articulate a generic CTA strategy.
Governments have tried to help develop so-called alternative technologies. In the
energy sector, the development of alternatives such as solar and wind energy were
subsidized. Other examples are production technologies developed to enhance workers'
qualifications and technology for the handicapped. In all these cases, some successes
have been achieved. At the same time, the risk of this "support and subsidy route" has
become clear. It can result in technologies with limited societal robustness and/or
viability in the marketplace. They continue to need to be protected from the wider
world. Governments developing desired technological options themselves may end up
with second-rate technology. It can only become competitive through exposure to
increasingly demanding environments (which is difficult for a public agent) and/or
mobilization of firms to produce such technologies. Immediate privatization and market
testing is not a solt~tion, however. An alternative technology needs some protection, a
safe niche in an otherwise too harsh selection environment. Too much protection, on
the other hand, will only create expensive failures.
Governments developing and introducing new technologies aimed at solving identified problems thus face major dilemmas that need to be recognized as such in order to
mitigate them. The key step is ensuring the development of a technology that will be
able to survive without government support. The question then is how to ensure a viable
technology is developed through a process of learning and constructive assessments and
reassessments of the various actors involved. Elsewhere, we have developed the concept
of strategic niche management to describe such a process in general terms [40, 44].
Strategic niche management can be defined as the orchestration of the development
and introduction of new technologies through setting up a series of experimental settings
(niches) in which actors learn about the design, user needs, cultural and political acceptability, and other aspects.
To use strategic niche management as a CTA strategy, the learning component
must take precedence over the goals of the technology actor. Learning must be broad
(explore linkages) as well as deep (involve first and second-order learning). In the
examples studied so far, first-order learning was dominant; learning was limited to a
detailing of specific design requirements, user needs, or social acceptance. In addition,
these aspects were seen as separate aspects and were taken up sequentially; that is, the
technical was optimized before the social and other aspects were considered [29, 45].
262
J. SCHOT AND A. RIP
Experiments with new technologies occur, and have become regular phenomena in
some sectors. But they are often not used to learn about possible new linkages between
technology, demand requirements and issues of cultural and political acceptability, nor
are such insights fed back into the strategies of actors and used on future occasions.
More action research applying CTA criteria is necessary to develop this generic
strategy further. The key point is that anticipation of later phases in the process of
graded introduction links the actor, who at first only wants to be successful in introducing
a new technology, to broader societal issues and thus forces him to broaden his design
and development process.
LOCI FOR REFLEXIVITY AND FEEDBACK: ALIGNMENT
With the first two generic strategies, the co-evolutionary dynamic of technology
and society was modulated from the demand or societal side, and from the supply or
technology side, respectively. A third generic strategy (not completely separable from
the other two) focuses on the interactions as such and attempts to create and exploit
loci: actual spaces, forums, and institutionalized linkages between supply and demand
(or variation production and selection environment) offering opportunities to modulate developments.
Several opportunities exist already. First, in the context of TA and CTA actions
forums (dialogue workshops, consensus conferences, etc.) have been used as instruments.
The process of participating in such forums stimulates anticipation, learning (both in a
broad and in a deep sense), and reflexivity. However, most often, these forums are
temporary loci, and they are used in a context distant from technology development.
Feedback is limited, and the outcomes have little force by themselves.
Second, in the context of technology policy and strategies for firms to get their
new products accepted (for example for high definition TV, telework, and specific
biotechnology applications), new platforms are created. Evaluations of these platforms
have shown, however, that anticipation is often short circuited when attempting to
identify the barriers to introduction, while feedback assists in attempts to remove
these barriers. Having other actors involved (trade unionists, consumer representatives,
environmentalists) is important and occurs in practice, even if primarily for legitimation
purposes. Learning is restricted to first-order learning and does not include an exploration of new linkages.
Third, over time regular nexusses have developed between variation and selection,
or supply and demand; one example is how test labs and trials have become an accepted
activity of the technology developers, rather than sending new products out into the
selection environment immediately [31]. Another example is environmental departments
within firms, which introduce (or could introduce) environmental criteria into design
processes [34]. When a nexus has become institutionalized, learning shifts from the
issue of how variation and selection can be linked, to learning how to handle specific
technologies within the nexus. The institutionalization of the nexus makes it forceful,
but it may also create barriers to further broaden the design and development processes.
To be able to use the platforms, dialogue workshops and networks and other
nexusses as a generic CTA strategy, further steps are necessary. The present examples
are either too distant from design or too much a part of technology-push introduction
strategies. To make them really effective for CTA, they should function as negotiation
spaces in the sense of Staudenmaier's [46] critical analysis of modern technology as
increasingly being developed in top-down systems approaches which push the alternative
approach--negotiation--to the margins.
C O N S T R U C T I V E T E C H N O L O G Y ASSESSMENT
263
What becomes clear from this argument is that CTA strategies have to function
in modern societies which have their own dynamics, and will thus depend on these
dynamics for their success. On the other hand, these dynamics need not remain the
same, and CTA activities are one input into processes of societal change.
The Future of CTA: Management of Technology in Society
The generic CTA strategies remain close to specific technologies, and they are
often developed in an instrumental way. This may well make them effective in the short
run, but as we have noted, there is more at stake. CTA activities are adopted in the
hope of contributing to the realization of better technologies (in a better society), but
this is embedded in the way modern societies handle technology more generally. Technical change is not perceived as a negotiable process. Actors often try to reduce space
for negotiating to the direction and nature of technical change [46]. Negotiation, if it
occurs, is forced into a storyline of proponents versus opponents [36]. Elsewhere, we have
discussed the two-track regime of separation of promotion and control of technology, for
example in the division of labor between government agencies [6].
The ultimate success of CTA strategies, therefore, depends on how our societies
evolve, specifically whether more appreciation and space will emerge for broader negotiation processes, and whether sociotechnical criticism will be appreciated as an essential
contribution in a society which has bound its fate to technology.
By way of contrast, it is interesting to look briefly at the situation in the early 19th
century, when technology became an explicit component of modern society. At the
time, there was sociotechnical criticism, carried by the Luddites--which by now have
become stereotypical figures symbolizing unreasoned protest against new technology.
Luddites--named for their legendary leader Ned Ludd--were textile workers who
resisted the introduction of new machines, eventually through machine smashing. In
contrast to the stereotypical view, we see Luddites as seeking to direct technology
toward societal purposes and avoid negative impacts. There is much historical evidence
that Luddites were neither anti-technology nor workers who were fighting against any
novel development [47-49].4 The Luddites judged machines on their contribution toward
the quality of labor and society as a whole, as if they were constructive technology
assessors avant la lettre. Luddite protests and actions were not an isolated happening.
Such activities were common practice in early modern Europe and were allowed by
common law and legal rules. The Luddites' protest was not directed against machines
or innovations in general, but it was taken up only where the introduction of new
machinery threatened living standards, conditions of labor, quality of products, or craft
status. Their interest, we could say, was not in the machines, but in the loss of access
to and negotiation power about new technologies.
Do we need neo-Luddism? Not in a sense of smashing machines. The aim of CTA
is to build better machines. However, Luddism, in the sense of sociotechnical criticism,
remains important. It should not take the form of sabotage perhaps (although we do
not condemn that out of hand). But the basic CTA philosophy does appear to have an
element of Luddism. It criticizes as too limited the sequential approach of optimizing
the technical before considering uptake, use, and effects, and it advocates broadening
Actors taking an anti-technology stance may use this negative image of Luddites as a honorary nickname
[52]. See also The Financial Times, April 15, 1996, reporting about the second Luddite Congress.
264
J. SCHOT AND A. RIP
design criteria of the kind of actors involved. There are important stakes involved in
neo-Luddite protest: to create and maintain space for sociotechnical criticism.
Why have we discussed Luddism at some length? It allows us to locate CTA in a
longer history of challenges to an emerging, and after some time dominant, two-track
regime of managing technology in society. In the two-track regime, promotional activities
are separated from control and regulation. Our institutions are set up in this way,
with regulatory agencies separate from technology-promotional agencies. Freedom of
innovation is institutionalized in specialized academies and laboratories, while the political regulation of technology is restricted to only the most certain and negative consequences. Separating promotion and control effectively decoupled the multiple constituencies of a technology. Thus, promotion actors (engineers and others) need not realize
that when they are engineering technology they are also engineering society. Many of
the negative or undesirable outcomes of the activities of the promoters derive from
this lack of insight. As we noted before, because promoters think their business is to
focus on the technical, they redefine the social as barriers to be overcome and approach
it as what they must do to make the technical successful. In this way, it becomes
fortuitous if the outcome is sociotechnically optimal.
Another effect of the modern two-track regime is the dominant ideology that
technological development cannot and should not be resisted. Increasingly, the broad
concept of social progress became identified with a notion of technological progress,
and Western societies came to see their emerging technological superiority as evidence
of an overarching cultural superiority. While this diagnosis has been made before, and
the ideology has been criticized and challenged, our point here is the difficulty of
actually overcoming the disadvantages of the two-track regime. The limited impact of
the Tennessee Valley Authority's comprehensive sociotechnical planning, the regionalist
movement associated with Lewis Mumford, and the alternative technology movement
in the 1960s are indicators of the difficulties.
While there is a certain rationale to the separation of promotion and control, and
we certainly do not want to advocate abolition of either or both tracks, we do conclude
that it reduces opportunities for learning, anticipation, and reflexivity. Conversely, if
CTA is applied widely, the two-track regime will be modified beyond recognition.
It is interesting to speculate about what could happen. In fact, if we view CTA as
a social technology that is now being introduced, it is essential to anticipate according
to our own perspective. One effect of different actors adopting CTA, and its becoming
a regular part of societal practices, is that the role of government in managing technology
in society will diminish. Agenda building as well as arrangements for better technology
will occur at lower levels. Governments would have a role in monitoring the quality of
societal learning in such situations. A further, and politically important, role of government as "overseer" is to ensure a balance in the access and voice of various constituences
involved in technical change (which might include support to develop the competence
of members of a constituency). This is not just a general political stance of ours. The
decentralization of management of technology in society made possible by CTA can
easily lead to a situation of technological neo-corporatism, with its own rigidities.
Conclusion
As we illustrated in the second and third sections, CTA agents, as well as other
bodies, organizations, firms, and societal groups, can be institutions charged with a TA
mission. There is a tension derived from the action perspective of CTA. On one hand,
CONSTRUCTIVE TECHNOLOGY ASSESSMENT
265
CTA agents are modulators of various kinds: mobilizers, advisers, mappers of coproduction dynamics, change agents. On the other hand, to actually exert influence
entails becoming an actor oneself, or at least a forceful visitor.
TA institutions feel this tension, especially when they want to emphasize CTA.
For the OTA, it was important to maintain an image of objectivity, of standing above
partisan struggles (even if this did not guarantee its institutional survival). For some of
its European counterparts, especially in the Netherlands and Denmark, a measure of
action and interference was accepted, as long as due processes were followed; for
example, no exclusion of particular views and particular parties. These differences, as
well as the common experience of the tension, derive from present political cultures.
These cultures may evolve and accommodate more negotiation. From the point of view
of actual feedback into design and development of technologies, one might even welcome
controversies. It has been demonstrated that TA reports have a larger impact when
there is a controversy involved [50, 51].
Deeper questions related to what we call the value profile of C T A agents cannot
be avoided. One such question is about representation. Who represents actual or potential users, victims or impactees in CTA activities, and how? Questions of representation
become even more tangled when future generations should be taken into account. At
a practical level, very important for productive CTA practices, institutional questions
have to be resolved: what sort of accountability can and must be organized for CTA
agents? For TA institutions, there is a mission and some form of public accountability.
For other CTA agents, say nongovernmental organizations and environmental groups,
the situation is more diffuse.
CTA agents are intermediaries between a future better world and the present
situation. Therefore, the issue of representation is easier when there is a broad consensus,
even when this is a consensus about diffuse goals, such as environment, and more
recently, sustainable development. CTA agents can assume the validity of such a goal
and proceed to act on that basis. It is indicative that CTA activities with an emphasis
on environmental aspects are the ones which have gone farthest in the direction of
feedback and implementation.
If there is no consensus, a CTA agent has to accept responsibility for the goals
embedded in its action. In the case of modern biotechnology, to mention one example,
this requires hard choices: about possibilities and risks of technological progress, and
about the legitimacy of the public concerns and their spokespersons. Accepting such a
responsibility can be interpreted as choosing sides, and one can, indeed, not avoid doing
so. But there should always be an argument in terms of anticipation, societal learning,
reflexivity, and goals at a meta level that justifies the agent's orientation and strategy.
Does this amount to a definable role and value profile of a CTA agent? To some
extent, yes. CTA agents embrace values such as being anticipatory, reflexive, and
oriented toward learning. While moral entrepreneurship for particular causes is widespread, characteristic for CTA is the attitude of making these causes part of an articulation strategy. Because of the latter, CTA agents have a moral thrust and moral justification of their being active as a CTA agent. Thus, they are also moral entrepreneurs, but
on a meta level. This is very visible in the emphasis on process (in actual CTA activities
and in the three generic strategies we outlined). The value accorded to articulation
explains (and contains an argument) why open-ended societal learning processes, especially if these contain concrete technologies and experiments, are emphasized by
CTA analysts.
266
J. SCHOT A N D A. RIP
CTA agents, however, will also act from their own perspective and interests. The
Ludds of this world are often not interested in open-ended learning processes, but in
pushing their perspective. Technology promoters are usually interested in CTA only
as a matter of prudence. We have argued already that such dichotomies might lessen
over time. Joint agenda building is what CTA agents contribute to, even when they act
primarily for their own interests. Thus, opposition to technology, including contestation
and harassment, is useful if it contributes, directly or indirectly, to the improvement of
technology in society. It puts on the agenda the question that there might be problems
on the horizon and that something may be done about it.
Our analysis of new responsibilities in managing technology in society might be read
as an argument for everyone to share the value profile of the CTA agent. Industrialists as
well as neo-Luddites are actors in CTA, whether the label is used or not, and they
should accept the values involved. But there is an irony involved. If actors would identify
completely with the open-ended learning values of intentional CTA, this would decrease
the stakes in the interaction, and thus the incentives to learn (because it is much easier
not to have to learn).
We can rephrase this as a dilemma for the future of CTA. On one hand, there is
a notion of shared responsibilities for managing technology in society, with all actors
working toward the CTA goals of learning, reflexivity, and anticipation. On the other
hand, there are the struggles between opponents of various kinds that are necessary to
have incentives to act and learn. Between the horns of this dilemma is the real world
and its interdependencies between actors.
If actors accept this irony and develop a competence in coping with it when coproducing technology and its effects, and when co-producing learning about such processes and the goals involved, there need be no harmony. But there will be reflexivity,
and a real chance to achieve better technology in a better society.
References
1. Smits, R., Leyten, J., and Den Hertog, P.: Technology Assessment and Technology Policy in Europe: New
Concepts, New Goals, New Infrastructures, Policy Sciences 28, 271-299 (1995).
2. Grin, J., and Graaf, H. van der: Technology Assessment as Learning, Science, Technology and Human
Values 21, 72-99 (1996).
3. Cronberg, T.: Technology Assessment in the Danish Socio-Political Context, Assessment Texts No. 9, The
Unit of Technology Assessment, Technical University of Denmark, Copenhagen, 1991
4. Ministerie van Onderwijs en Wetenschappen: lntegratie van Wetenschap en: Technologie in de Samenleving,
Beleidsnota, 's-Gravenhage: Tweede Kamer 1983-1984, 18 421, nrs. 1-2 (Policy Memorandum: Integration
of Science and Technology in Society).
5. Daey Ouwens, C., Hoogstraten, P. van, Jelsma, J., Prakke, F., and Rip, A.: Constructief Technologisch
Aspectenonderzoek. Een Verkenning, Den Haag: Staatsuitgeverij (1987) (NOTA Voorstudie 4).
6. Rip, A., Misa, T. J., Schot, J. W. (eds.): Managing Technology in Society. The Approach of Constructive
Technology Assessment, Pinter Publishers, London, 1995.
7. Rathenau Instituut: Van Technologisch Aspectenonderzoek tot Maatschappelijk Debat, 10 Jaar NOTARathenau lnstituut 1986-1996, Den Haag, 1996.
8. Andersen• •. E. (ed.): Feasibi•ity Study •n New Awareness •nitiatives• Danish B•ard •f Techn•••gy•
C•penhagen, 1995.
9. Rathenau Institute: The Rathenau Institute and the Debate, Annual Report 1994, Den Haag, 1991.
10. NOTA: Technology Assessment: To Adjust or to Channel, Den Haag, 1994.
11. Dieleman, H., and Hoo, S. de: Understanding Technological Responses of Industrial Firms to Environmental Problems: Implications for Government Policy, in Environmental Strategies for Industry, International
Perspectives on Research Needs and Policy Implications. K. Fischer and Johan Schot, eds., Island Press,
Washington, D.C. and Covelo, CA, 1993.
CONSTRUCTIVE TECHNOLOGY ASSESSMENT
267
12. Schot, J. W.: Technology Dynamics: An Inventory of Policy Implications for Constructive Technology
Assessment, in Maatschappelijke Sturing van Technische Ontwikkeling. Constructief Technology Assessment
als hedendaags Luddisme. J. W. Schot, Ph.D. thesis, Universiteit Twente, Enschede, 1991.
13. Jelsma, J., and Rip, A., with a contribution by J. L. van Os: Biotechnologie in Bedrijf. Een bijdrage van
Constructief Technology Assessment aan biotechnologisch innoveren, The Hague: Rathenau Institute, 1995.
Report of a Project to Elaborate Constructive TA for Firms in the Area of Biotechnology. English summary
available (Biotechnology in Business).
14. Fonk, G. J.: Een Constructieve Rol van de Consument in Technologie-ontwikkeling. Constructief Technologisch Aspectenonderzoek Vanuit Consumentenoptiek, SWOKA, The Hague, Ph.D. thesis, University of
Twente, 1994.
15. Fonk, G. J., and Hamstra, A.: Toekomstbeelden voor Consumenten van Novel Protein Foods, Delfl.
Programma DTO, werkdocument VN12, 1996.
16. Vergragt, P. and Nout, D. van: Introducing the Mobile Hydrogen Fuel Cell, paper presented at the Fourth
Greening of Industry Network Conference, Toronto, November 12-14, 1995.
17. Ministerie van Economische Zaken: Technologie en Samenleving, Den Haag, 1991.
18. Ministerie van Economische Zaken: Kennis in Beweging, Den Haag, 1995.
19. Remrnen, A.: Constructive Technology Assessment, in Danish Experiment: Social Constructions of Technology. T. Cronberg et al., eds., New Social Science Monographs, Copenhagen, 1991.
20. Remmen, A.: Pollution Prevention, Cleaner Technologies and Industry, in Managing Technology in Society:
The Approach of Constructive Technology Assessment. A Rip, Th. J. Misa, and J. W. Schot, eds., Pinter
Publishers, London, 1995, 199-222.
21. Hetland, P.: Exploring Hybrid Communities: Telecommunications on Trial. Ph.D. thesis, Hedmark Collegc,
Norway, 1994.
22. Herbold. R.: Technologies as Social Experiments: The Construction and Implementation of a HighTech Waste Disposal Site, in Managing Technology in Society. The Approach of Constructive Technology
Assessment. A, Rip, T. Misa, and J. Schot, eds., Pinter Publishers, London, 1995, 185-197.
23. OECD: New Technologies in the 1990s: A Socio-Economic Strategy, OECD, Paris, 1988.
24. Carnegie Commission: Enabling the Future. Linking Science and Technology to Societal Goals. Carnegie
Commission, New York, 1992.
25. Branscomb, L, M.: Empowering Technology. Implementing a U.S. Strategy. MIT Press, Cambridge, MA,
1993,
26. Soete, L.: (Constructive) Technology Assessment: An Economic Perspective, in Managing Technology in
Society. 7"he Approach of Constructive Technology Assessment. A. Rip, T. J. Misa, and J. W. Schot, eds.,
Pinter Publishers, London, 1995, 36-49.
27. Smits, R., and Leyten, J,: Technology Assessment: Op Zoek Naar een Bruikbare Aanpak. 1. Mogelilkheden
on Beperkingen, Staatsuitgeverij, Den Haag, 1984. (Background Study to the Policy Memorandum) lntegratie van Wetenschap en Technologie in de Samenleving.
28. Rip, A., Belt, H. van den, and Schwarz, M.: Theoretisehe Analyses, in ('onstructiefTechnologisch Aspectenonderzoek. Een Verkenning. (NOTA Voorstudie 4) C. Daey Ouwens, P. van Hoogstraten, J. Jelsma, F.
Prakke, and A. Rip, eds., Staatsuitgeverij, Den Haag, 1987.
29. Deuten, J., and Rip, A.: R&D Management for Public Acceptance: An Important Aspect of Total Quality.
Dept. Philosophy of Science and Technology, University of Twente. Enschede, 1996. Paper presented to
the International Conference on R&D Management, University of Twente, March 1996.
30. Den Hond, F,, and Groenewegen, P.: Environmental Strategies for Industry, International Perspectives on
Research Needs and Policy Implications. K. Fischer and Johan Schot, eds., Island Press, Washington. D.C.
and Covelo, CA, 1993.
31. Belt, H. van den, and Rip, A.: The Nelson-Winter/Dosi Model and Synthetic Dye Chemistry, in 7'fie Social
Construction of Technological Systems. New Directions in the Sociology and History of Technology. Wiebe
E. Bijker, Thomas P. Hughes, and Trevor J. Pinch, eds.: MIT Press. Cambridge, MA, 1987.
32. Bijker, W. E., and Law, J.: Shaping Technology~Building Society: St,dies in Sociotechnical Chatlge. M IT
Press, Cambridge, MA, 1992.
33. Rip. A., and Kemp, R.: Towards a Theory ~fSocio-Technica/Change, manuscript, University of Twente,
August 1996, 85.
34. Schot, J. W.: De lnzet van Constructief Technology Assessment, Kennis en Methode 3~ 265-293 (1996).
35. Wynne, B.: Technology Assessment and Reflexive Social Learning: Observations from the Risk Field, in
Managing Technology in Society. The Approach of Constructive Technoh)gy Assessment. A. Rip, T. J.
Misa, and J. W. Schot, eds.: Pinter Publishers, London, 1995, 19-36.
36. Rip, A., and Talma, S.: Antagonistic Patterns and New Technologies, in Coordinating Socio-Technical
Orders. C. Disco and B. J. R. Van der Meulen, eds., Walter de Gruyter, Berlin, forthcoming 1997.
268
J. SCHOT AND A. RIP
37. Akrich, M.: User Representations: Practices, Methods and Sociology, in Managing Technology in Society.
The Approach of Constructive Technology Assessment. A. Rip, T. J. Misa, and J. W. Schot, eds., Pinter
Publishers, London, 1995, 167-185.
38. Mintzberg, H.: The Rise and Fall of Strategic Planning, Prentice-Hall, New York, 1994.
39. Schot, J. W.: Constructive Technology Assessment and Technology Dynamics: The Case of Clean Technologies, Science, Technology & Human Values 17, 36-56 (1992).
40. Schot, J. W., Hoogma, R., and Elzen, B.: Strategies for Shifting Technological Systems. The Case of the
Automobile System, Futures 26, 1060-1076 (1994).
41. White, L. J.: The Regulation of Air Pollutant Emissions from Motor Vehicles, American Enterprise Institute,
Washington, D.C., 1982.
42. White, L. J.: U.S. Mobile Source Emission Regulation: The Problems of Implementation, Policy Studies
Journal 11, 77-87 (1982).
43. Rip, A.: Introduction of New Technology: Making Use of Recent Insights from Sociology and Economics
of Technology, Technology Analysis & Strategic Management 7(4), 417--431 (1995).
44. Rip, A.: Between Innovation and Evaluation: Sociology of Technology Applied to Technology Policy
and Technology Assessment, RISESST 2, 39--67 (1992). (Also published in Italian: 'Tra Innovazione e
Valutatione. La Sociologia Applicata alia Politica Ed alia Valutaziome della Tecnologia,' in Leonardo
Cannovb (cur.), Studi Sociali della Tecnologia. Metodologie Integrate di Valutazione. Euroma/La Goliardica,
Roma, 1991, pp. 63-105.
45. Hoogma, R., and Schot, J. W.: How Innovative Are Users? A Critique of Learning-by-Doing and Using.
Paper presented at 4S/EASST conference, Bielefeld, 10-13 October (1996).
46. Staudenmaier, J. M.: The Politics of Successful Technologies in In Context: History and History of Technology. Essays in Honour of Melvin Kranzberg. R. C. Post, and S. H. Cutliffe, eds., Lehigh University Press,
Bethlehem, PA, 1989, 150--171.
47. Hobsbawn, E. J.: The Machine-Breakers, in Labouring Man. E. J. Hobsbawn, Weidenfeld & Nicholson,
London, 1964.
48. Thompson, E. P.: The Making of the English Working Class, Victor Gollancz, New York, 1963.
49. Randall, A.: Before the Luddites. Custom, Community and Machinery in the English Woolen Industry,
1776-1809, Cambridge University Press, Cambridge, 1991.
50. Rip, A.: Societal Processes of Technology Assessment, in Impact Assessment Today. H. A. Becker and
A. L. Porter, eds., Uitgeverij Jan van Arkel, Utrecht, 1986, 415-434.
51. Whiteman, D.: Congressional Use of Technology Assessment, in The Politics of Technology Assessment,
Institutions, Processes and Policy Disputes. D. M. O'Brien and D. A. Marchand, eds., Lexington Books,
Lexington, MA, 1982, 51~54.
52. Sale, K.: Rebels Against the Future. The Luddites and Their War on the Industrial Revolution. Lessons for
the Computer Age, Addison-Wesley, Reading, MA, 1996.
53. Schwarz, M., and Thompson, M.: Divided We Stand. Redefining Politics, Technology and Social Choice,
Harvester Wheatsheaf, New York, 1990.
Received 31 July 1996; accepted 23 October 1996